Access and Cost Barriers to Mental Health Care by Insurance Status, 1999 to 2010

Abstract

The cost of mental health services has always been a great barrier to accessing care for people with mental health problems. This article documents changes in insurance coverage and cost for mental health services for people with public insurance, private insurance, and no coverage. Compared to 1999–2000, in 2009–10 people with mental health problems were more likely to have public insurance and less likely to have private insurance. Although access to specialty care remained relatively stable for people with mental illness, cost barriers to care increased among the uninsured and the privately insured who had serious mental illness. The rise in cost barriers among those with private insurance suggests that the current financing of care in the private insurance market is insufficient to alleviate cost burdens and has implications for reforms under the Affordable Care Act. People with mental health problems who are newly eligible to purchase private insurance under the act might still find high cost barriers to accessing care.

Common mental illnesses such as depression can be extraordinarily disabling, yet many people with those illnesses do not receive treatment. In fact, three out of five adults with a recent mental health disorder did not receive care from either a general medical provider or a mental health specialist.¹ It is an oversimplification to suggest that all of these people needed treatment.² However, even many people with serious disorders do not receive mental health care.³

Financial barriers are one impediment to receiving needed care. People often cite concerns about the cost of care or lack of health insurance coverage as reasons for not receiving mental health care.^4,5 In the National Comorbidity Study, for example, 47 percent of respondents with a mood, anxiety, or substance-use disorder who said they thought they needed mental health care cited cost or not having health insurance as a reason they did not receive that care.⁴ The percentage of people who forgo mental health care because of its cost may also be increasing.⁵

Insurance Coverage, Costs, and Access to Care

People with mental illness are less likely to have health insurance than those without mental health problems.^3,6–8 Using the Medical Expenditure Panel Survey for 2004–06, for example, Rachel Garfield and colleagues found that 37 percent of working-age adults with severe mental illness were uninsured for at least part of the year, compared to about 28 percent of people without severe mental illness.⁸ Even after controlling for demographic differences, William Pearson and colleagues found that the odds of having health insurance were 40 percent lower for people with serious psychological distress (SPD) than for those without SPD in 2007.⁶

For low-income adults and disabled people, eligibility for Medicaid provides some protection against financial barriers to care. Indeed, Medicaid is the largest single payer for mental health services in the United States.⁹ In 2010 approximately 33 percent of the adults receiving Medicaid met the criteria for having had a mental illness in the past year, and 11 percent met the criteria for having had a serious mental illness.⁹

The role of insurance coverage in increasing the use of care depends on the severity of the mental illness assessed and the type of service used.^8,10,11 Evidence from the National Comorbidity Survey Replication suggests that among people with a mental health disorder, the insured are more likely to use the health care sector, while the uninsured are more likely to use human services, complementary or alternative medicine, and the like.¹ Other researchers have found that rates of mental health care for people with severe mental illness are lowest for the uninsured and highest for those with public insurance, while those with private insurance fall between the other groups.^3,8

Even among the insured, costs may be a barrier to getting needed mental health care. Cost sharing may disproportionately affect people with mental illness, who have lower family incomes and are more likely to be living in poverty than those without mental illness.^12,13

Recent trends in the insurance market may exacerbate cost barriers to mental health care. Overall, there has been a drop in employer-sponsored health insurance coverage. Between 1999 and 2009 the share of firms offering coverage declined from 59 percent to 56 percent, and the share of workers with employer-sponsored coverage declined from 69 percent to 61 percent.¹⁴ The share of nonelderly people with public coverage rose from 12 percent to 17 percent during the same period.¹⁴

It is not clear whether these general trends in insurance coverage apply both to people with mental illness and to those without it. However, Roland Sturm and Kenneth Wells observed a decline in private coverage for people with mental illness in the late 1990s.¹⁰ And Sherry Glied and Richard Frank found that among people with an activity limitation due to mental illness, the share with private insurance declined from more than 50 percent to less than 42 percent between 1996 and 2006.¹⁵

There is also a continuing movement toward increased cost sharing for people with insurance.¹⁶ Those with mental disorders have substantial out-of-pocket expenditures for medical care, accounting for about 29 percent of mental health and substance abuse outpatient costs nationally.¹⁷ Approximately 14 percent of working-age patients with a mental disorder have out-of-pocket expenditures that exceed 20 percent of family income annually.¹⁸

The Deficit Reduction Act of 2005 gave states greater flexibility in charging premiums, copayments, or deductibles to people enrolled in their Medicaid programs. Coupled with the pressure on state budgets as the results of the Great Recession, which officially lasted from the end of 2007 to 2009, this flexibility led many states to increase cost sharing for their Medicaid beneficiaries.¹⁹

HEALTH POLICY

Recent years have also seen numerous state policy changes that affect care for persons with mental illness, such as mandates that insurers provide mental health benefits and the expansion of managed care to the disabled Medicaid population.^3,19 Such state specific arrangements results in substantial variations across states in coverage for mental health care.¹⁹ While provisions of the Mental Health Parity and Addiction Equity Act of 2008 can reduce out-of-pocket cost burdens, these provisions apply only if the insurer chooses to provide coverage for mental health and substance abuse disorders, and the act exempts certain categories of employers, such as small employers.²⁰

Under the Affordable Care Act, an estimated 32.1 million Americans will gain access to health insurance that includes a mental health and substance use benefit.²¹ States have the option to expand Medicaid to all people with incomes at or below 138 percent of the federal poverty level, a population at greater risk than more affluent populations risk for mental health problems.

In addition, all states will have either a state-based or federally facilitated health insurance exchange that will provide options for people purchasing health coverage in the individual and small-group market. Rating rules prohibit setting premiums based on preexisting conditions (including mental illness and substance abuse) or sex, and the rules strictly limit the ability of insurance carriers to set premiums based on age. People meeting the income eligibility criteria will be able to obtain premium subsidies in the form of a tax credit to help pay for coverage offered through an exchange and additional subsidies to help cover their cost sharing. ²²

Mental health and substance abuse services are among the ten essential health benefits that must be covered by insurance offered in the exchanges and by all new products sold in the individual and small-group markets. Furthermore, insurers are required to cover these services at parity with medical and surgical services.

Given the potential of the Affordable Care Act to substantially change insurance patterns and access to care for people with mental health problems, it is timely to examine the association between insurance and access to care for such people. We investigated changes in the patterns of insurance coverage for those with mental health problems. We also examined access to specialty care and perceived cost barriers among the privately and publicly insured and the uninsured.

Study Data And Methods

The data for this study came from the Integrated Health Interview Series (IHIS), an online database of more than forty years of data from the National Health Interview Survey. The IHIS is designed to facilitate the comparability of survey questions and improve access to survey data files.²² The survey is an annual household interview administered to a nationally representative sample of the noninstitutionalized civilian population, covering 75,000–100,000 people per year. Our analysis was limited to working-age adults—ages 18–64—who were interviewed between 1999 and 2010. Sample sizes over the period varied from 23,716 in 2008 to 31,741 in 2003. Response rates for the sample during this period averaged 69.4 percent, ranging from 60.8 percent in 2010 to 74.3 percent in 2002. We combined every two years’ worth of data to increase the precision of the estimates.

The dependent variables were the use of specialty care and cost barriers to care. The use measure was based on whether the respondent had seen “a mental health professional, such as a psychiatrist, psychologist, psychiatric nurse or clinical social worker” in the previous twelve months. The second measure assessed whether the respondent believed that he or she had needed mental health care or counseling but had not gotten it in the previous twelve months because he or she could not afford it.

The measure of mental health was based on two criteria. First, respondents were categorized as having SPD based on their scores for the six questions in the Kessler-6 scale, which assesses symptoms such as having feelings of sadness, hopelessness, and worthlessness in the previous thirty days. ²⁴ The scores were summed, and a total score of 13 or higher (out of a possible 24) was used as a dichotomous indicator for SPD. In addition, respondents were asked if depression, anxiety, an emotional problem, or another mental problem was the cause of a functional or activity limitation such as problems walking, standing, or participating in social activities. Responses on the two measures were used to divide the respondents into the following three mutually exclusive categories: no mental health problem (low SPD and no limitation due to a mental health problem), moderate mental health problem (either SPD or a limitation), and serious mental illness, (both SPD and a limitation).

Insurance status refers to the respondent’s coverage at the time of the interview. Coverage was categorized as private (employer-based, MediGap, or TRICARE—the Department of Defense’s health care program), public (Medicaid, Medicare, other government programs, or other military coverage), and uninsured. People with both private and public coverage were classified as private.

We tested for trends in mental health status within each insurance group and for changes in insurance coverage by mental health status. Tests for trends were computed using logistic regressions with time (measured as single years) as the independent variable. Full regression results and standard errors of all point estimates are provided in the online Appendix.²⁵ In addition, we computed trend analysis, controlling for sociodemographic characteristics and health status; these results are also available in the Appendix.²⁵

Estimates were weighted to represent the population. We used the statistical software Stata, version 12.0, to adjust standard errors to account for the complex sampling design.

Study Results

The percentage of the working-age population classified as having moderate mental health problems (SPD or a limitation) increased significantly, from 3.7 percent in 1999–2000 to 5.1 percent in 2009–10 (Exhibit 1). The percentage with serious mental illness (SPD and a limitation) also increased, from 0.9 percent to 1.4 percent.

Exhibit 1.

Percent Of People Ages 18–64 With Mental Health Problems By Insurance Status, 1999–2010

Insurance status (n)	1999–2000	2001–02	2003–04	2005–06	2007–08	2009–10
None
SPD or limitation (3,193)**	5.1	5.7	5.5	5.1	5.8	6.1
SPD and limitation (791)***	1.0	1.3	1.4	1.3	1.3	1.6
Public
SPD or limitation (4,642)	14.5	16.2	13.7	14.9	15.2	15.4
SPD and limitation (1,909)	5.5	6.8	6.6	6.1	5.8	5.5
Private
SPD or limitation (5,160)**	2.5	2.7	2.7	2.6	2.5	3.1
SPD and limitation (987)**	0.4	0.5	0.5	0.5	0.6	0.5
Total
SPD or limitation (13,104)****	3.7	4.3	4.2	4.3	4.5	5.1
SPD and limitation (3,723)****	0.9	1.1	1.2	1.2	1.3	1.4

SOURCE Authors’ analysis of data from the Integrated Health Interview Series, 1999–2010 (see Note 22 in text).

NOTES Subjects were divided into three mutually exclusive groups: those with no mental health problem, those with a moderate mental health problem (either serious psychological distress (SPD) or an activity or functional limitation due to a mental health problem), and those with a serious mental health problem (both SPD and a limitation). Private insurance is employer-based, MediGap, or TRICARE. Public insurance is Medicaid, Medicare, other government programs, or other military coverage. None is uninsured. People with both private and public coverage were classified as private. Significance indicators pertain to statistical tests for significance of trend.

^**p ≤ 0.05

^***p ≤ 0.01

^****p ≤ 0.001

Although the changes are modest, there was a significant increase in the proportion of the population with moderate mental health problems or serious mental illness among the uninsured and the privately insured (Exhibit 1). In every two-year period, the prevalence of mental health problems (both moderate and serious) was significantly higher among the publicly insured than among those in the other insurance categories, and it was lowest among the privately insured.

During the study period the percentage of people with moderate mental illness who had public coverage increased by approximately 34 percent (from 25.9 percent to 34.8 percent), and the percentage with private coverage fell by 21 percent (from 50.2 to 39.8 percent) (Exhibit 2). Although the rate of private coverage declined, and the rate of uninsurance and public coverage increased for those with serious mental illness (SPD and a limitation), these trends were not significant. The decline in private coverage was greater for people with any level of mental illness than for people with no mental health problems.

Exhibit 2.

Percent Of People Ages 18–64 By Mental Health And Insurance Status, 1999–2010

Insurance status (n)	1999–2000	2001–002	2003–04	2005–06	2007–08	2009–10
No mental health problem
None (53,395)****	17.4	17.4	19.1	19.8	19.6	21.1
Public (23,331)****	5.6	6.2	6.9	7.8	8.5	9.8
Private (178,024)****	77.0	76.4	74.0	72.4	71.9	69.1
SPD or limitation
None (3,193)	23.9	23.7	25.8	24.1	25.7	25.4
Public (4,642)****	25.9	29.0	27.0	32.8	34.6	34.8
Private (5,160)****	50.2	47.3	47.2	43.1	39.7	39.8
SPD and imitation
None (791)	19.9	20.3	23.3	21.5	20.7	26.3
Public (1,909)	43.5	45.5	45.5	47.5	45.3	47.3
Private (987)	36.6	34.2	31.2	31.0	34.1	26.4

SOURCE Authors’ analysis of data from the Integrated Health Interview Series, 1999–2010 (see Note 22 in text).

NOTES Subjects were divided into three mutually exclusive groups: those with no mental health problem, those with a moderate mental health problem (either serious psychological distress (SPD) or an activity or functional limitation due to a mental health problem) and those with a serious mental health problem (both SPD and a limitation). Private insurance is employer-based, MediGap, or TRICARE. Public insurance is Medicaid, Medicare, other government programs, or other military coverage. None is uninsured. People with both private and public coverage were classified as private. Significance indicators pertain to statistical tests for significance of trend.

^**p ≤ 0.05

^***p ≤ 0.01

^****p ≤ 0.001

In each two-year period, people with moderate mental health problems were significantly more likely to be uninsured than were people who had no problems. Rates of uninsurance did not differ significantly between those with no mental illness and those with serious mental illness, despite the greater need for services in the latter group.

Insurance coverage also varied significantly by level of mental health problems. In each time period, people with serious mental illness were less likely to have private insurance and more likely to have public insurance than people in the other two groups. During the study period 43.5–47.5 percent of people with serious mental illness were covered with public insurance, the most common form of coverage for this population.

For most insurance groups, there was no significant change over time in the probability of accessing specialty care (Exhibit 3). Among people with no mental health problems, there was a significant (albeit small—just over one percentage point) increase in the probability of accessing mental health care for the privately and publicly insured. Among people with moderate mental health problems and private coverage, there was a 2.4-percentage-point increase in the likelihood of using specialty care. The trends for people with serious mental health problems suggest reduced use of specialty care among the publicly and privately insured, but the trends were not significant.

Exhibit 3.

Percent Of People Ages 18–64 Who Saw A Mental Health Professional By Mental Health And Insurance Status, 1999–2010

Insurance status (n)	1999–2000	2001–02	2003–04	2005–06	2007–08	2009–10
No mental health problem
None (52,198)	3.4	3.6	2.7	3.1	3.1	3.2
Public (23,113)**	8.7	9.1	8.2	10.5	9.6	10.0
Private (176,548)****	4.9	5.1	5.3	5.3	6.0	6.4
SPD or limitation
None (3,152)	21.2	19.1	19.0	17.0	19.5	21.6
Public (4,571)	47.8	46.6	46.7	47.7	51.3	54.6
Private (5,113)***	33.9	31.1	31.8	34.3	38.7	36.3
SPD and limitation
None (784)	44.4	49.4	38.8	42.9	43.2	44.5
Public (1,888)	77.8	68.4	64.9	71.1	66.7	67.4
Private (980)	61.8	61.8	60.6	54.6	55.2	60.2

SOURCE Authors’ analysis of data from the Integrated Health Interview Series, 1999–2010 (see Note 22 in text).

NOTES Subjects were divided into three mutually exclusive groups: those with no mental health problem, those with either a moderate mental health problem (serious psychological distress (SPD) or an activity or functional limitation due to a mental health problem) and those with a serious mental health problem (both SPD and a limitation), Private insurance is employer-based, MediGap, or TRICARE. Public insurance is Medicaid, Medicare, other government programs, or other military coverage. None is uninsured. People with both private and public coverage were classified as private. Significance indicators pertain to statistical tests for significance of trend.

^**p ≤ 0.05

^***p ≤ 0.01

^****p ≤ 0.001

For each two-year period, among people with mental health problems, the uninsured had the lowest access to specialty care. On average during the study period, approximately 20 percent of those with moderate problems and 44 percent of those with serious problems accessed care. The publicly insured had the greatest access: Approximately 49 percent of those with moderate impairments and 69 percent of those with serious impairments received care. Across all three types of insurance, the use of services increased with the severity of the mental health problem.

The percentage of respondents with moderate mental health problems who reported that cost prevented them from getting needed mental health care increased over time for all insurance groups, although the increase was not significant for the publicly insured (Exhibit 4). For people with serious mental illness, cost barriers increased for the uninsured and the privately insured, although the trends were not significant. As shown in the Technical Appendix,²⁴ however, the trend for cost barriers to increase among the privately insured was significant in multivariate analysis.

Exhibit 4.

Percent Of People Ages 18–64Who Experienced A Cost Barrier To Mental Health Care By Mental Health And Insurance Status, 1999–2010

Insurance status (n)	1999–2000	2001–02	2003–04	2005–06	2007–08	2009–10
No mental health problem
None (52,931)***	3.8	4.1	4.8	4.7	5.4	5.0
Public (21,153)	2.2	2.2	2.2	3.2	2.7	2.4
Private (176,827)****	0.7	0.8	0.8	1.0	0.9	1.2
SPD or limitation
None (3,153)****	28.0	30.0	26.9	33.8	36.4	35.3
Public (4,581)	8.7	8.8	9.3	12.7	14.2	10.8
Private (5,125)***	8.6	9.3	10.0	12.0	13.3	12.6
SPD and limitation
None (784)	61.1	55.1	53.6	56.0	57.1	64.0
Public (1,887)	20.0	18.8	23.2	24.1	26.5	18.2
Private (981)	24.3	17.3	29.7	18.8	32.9	30.3

SOURCE Authors’ analysis of data from the Integrated Health Interview Series, 1999–2010 (see Note 22 in text).

NOTES Subjects were divided into three mutually exclusive groups: those with no mental health problem, those with a moderate mental health problem (either serious psychological distress (SPD) or an activity or functional limitation due to a mental health problem) and those with a serious mental health problem (both SPD and a limitation). Private insurance is employer-based, MediGap, or TRICARE. Public insurance is Medicaid, Medicare, other government programs, or other military coverage. None is uninsured. People with both private and public coverage were classified as private. Significance indicators pertain to statistical tests for significance of trend.

^**p ≤ 0.05

^***p ≤ 0.01

^****p ≤ 0.001

Not surprisingly, in each two-year time period, people with mental health problems (either moderate or serious) who were uninsured were more likely to report cost barriers to accessing care, compared to the privately or publicly insured. For example, in 2009–10, 64.0 percent of the uninsured with serious mental health problems reported a problem accessing care as a result of costs, compared to 18.2 percent of those with public insurance and 30.3 percent with private insurance. Within each insurance type, the probability of cost barriers also increased with the severity of mental illness.

Discussion

Consistent with prior studies, we found that people with mental health problems were more likely to be uninsured that those without such problems.^3,6–8 And consistent with trends in the general population, we found a decline in private insurance and an increase in public insurance for people with mental illness.

The decline in private coverage was greater among people with mental health problems than among those without such problems. The overall decline in private insurance was driven, in part, by the movement of workers into jobs that were less likely to offer insurance and a reduction in the take-up of employer-sponsored insurance among low- and middle-income workers because of the high cost of premiums.²⁶

Our results speak to the importance of public insurance in covering care for people with mental illness. Indeed, by 2009–10, 47.3 percent of those classified as having a serious mental illness were covered by public insurance, compared to only 9.8 percent of those with no mental health problems (Exhibit 2). The role of public coverage significantly expanded during the study period, especially for people with moderate mental health problems.

Public insurance typically has more generous benefits than private insurance for people with mental health problems. Across the study period we found that the publicly insured had greater access to mental health specialists than did the privately insured, a difference that reflected both the greater level of disability and the richer benefits of people with public insurance.

Previous work has suggested that access to specialty mental health services increased between 1997 and 2002.⁵ We found this trend only in the case of people with moderate levels of mental health problems who had private insurance. Indeed, for people with serious mental health problems who had either public or private insurance, the use of specialty care declined during the study period. However, those trends were not significant, perhaps because of our study’s small sample sizes and insufficient statistical power. This finding is consistent with what has been observed in prior research: Access to specialty care between 1996 and 2006 among those with limitations associated with a health condition has declined.¹⁵

Access to specialty care increased with the severity of mental disorder, an encouraging finding that reflects the important role of public programs. Nonetheless, in 2009–10, among people with the most serious level of mental health problems, about 33 percent of those with public coverage, 40 percent of those with private coverage, and 56 percent of those without insurance did not access care (Exhibit 3). Clearly, there are many factors that determine help-seeking behavior in illness, such as stigma, attitudes toward treatment, interpretations of mental health symptoms, and the availability of mental health professionals.² However, cost is a very important barrier, and it can be an impediment to care even among the insured.

Cost barriers increased between 1999–2000 and 2009–10 for people with moderate mental health problems who had private coverage or who lacked insurance. Cost barriers also increased for people with serious mental illness who had private insurance, although this trend was significant only in multivariate analysis. In addition, cost barriers were higher for people with serious mental illness across all insurance types, compared to those with moderate or no mental illness, which may reflect a need for more frequent or intensive services on the part of those with more serious illness.

Patients who cannot afford out-of-pocket costs may forgo treatment, leading to poor management of the mental illness. Although out-of-pocket costs are reported to have declined from 1986 to 2005 as a proportion of all mental health care spending, they are higher in terms of absolute dollars.⁹ Prescription drugs account for almost two-thirds of out-of-pocket spending for mental health care, and simply expanding insurance coverage would not reduce this potential barrier to treatment.¹⁷

Of course, cost sharing may reduce the potential problem of “unnecessary” mental health care use--that is, use in absence of substantial need-- by imposing a financial constraint. However, the fact that cost barriers are highest among those with serious mental health problems is a concern. In 2009–10, among people with serious problems, 18.2 percent of those with public health insurance and 30.3 percent of those with private insurance reported that costs were a barrier to getting needed mental health care (Exhibit 4). These rates were considerably higher than the rates for people with moderate mental health problems.

A key limitation of this study is that although we were able to divide the type of health insurance into the broad categories of public, private, and uninsured, we did not have specific information on the type of benefits in each plan and on whether mental health services were included. In addition, for people covered by Medicaid, the benefit structure—including out-of-pocket costs and whether the program had a type of limited benefits set offered through 1115 waivers—could make a difference in coverage and use. A more thorough assessment of the impact of particular plan provisions on access to services and costs is needed.

In addition, we assessed only specialty mental health care. Given that much mental health care takes place in the general health care sector, we underestimated the use of mental health services.

Conclusion

The good news is that the Affordable Care Act has expanded parity provisions to new plans and specified that mental health and substance abuse services are an essential health benefit. The act also expanded federal parity provisions to a broader set of health plans: Medicaid Alternative Benefit Plans and plans offered through the individual market.²⁰ However, self-insured plans are not included in the new coverage mandate for mental health and substance abuse services, and certain small employers are exempt when purchasing coverage in a health insurance exchange.²⁰

Overall, the act presents a great opportunity to expand health insurance coverage to people with mental health disorders. Garfield and colleagues estimate that, if fully implemented, the act could provide some 3.7 million adults with severe mental illness with coverage.⁸

Yet there are still major challenges ahead. The state Medicaid expansions are optional. As of August 2013, twenty-four states and the District of Columbia have decided to implement expansions, twenty-one states have decided not to do so, and five states are continuing to debate expanding their Medicaid programs.²⁷ The key group left behind in states opting not to implement the expansion are people who are currently ineligible for Medicaid—low-income adults without children—which is a group at greater risk for serious mental health illness than the general population.

For states that do not expand Medicaid, new rules issued by the Department of Health and Human Services indicate that low-income people will not be required to buy insurance. Moreover, even in states that do expand Medicaid, there is considerable flexibility in the benchmarks they adopt for defining coverage.²⁸ If states adopt Medicaid Alternative Benefit Plans and benchmark them in line with private insurance, coverage will be much less generous than in many existing Medicaid plans. Thus, people with mental health problems might gain access to care, but they might also face considerable cost barriers to that care.

Policy makers face additional challenges under health reform in coordinating the delivery of mental and behavioral health care across agencies and professionals, providing additional needed supports for people with mental health problems, and addressing the paucity of providers in the mental health specialties. Certainly the Affordable Care Act has great potential to provide coverage for currently uninsured people with mental health problems. But how well it will address access and cost barriers is unclear. The devil, of course, now lies in the details of the rules, regulations, and implementation of the act that will apply to people with mental health problems.